Prediction of Ontario Hourly Load Demands and Neural Network Modeling Techniques

Author

Findlay, Raymond ; Liu, Fang

Author_Institution

Power Res. Lab., McMaster Univ., Hamilton, Ont.

fYear

2006

fDate

38838

Firstpage

372

Lastpage

375

Abstract

Accurate and reliable load forecasting is necessary to ameliorate energy management. For the purpose of load demands prediction, this paper develops an artificial neural network model, which adopts Levenberg-Marquardt method as training algorithm, both visual comparison and statistical techniques as validation methods. With the built neural network model, the hourly load demands of Ontario in 2004 have been successfully forecasted

Keywords

learning (artificial intelligence); load forecasting; neural nets; power engineering computing; power system control; Levenberg-Marquardt method; Ontario hourly load demand prediction; artificial neural network modeling technique; energy management; load forecasting; statistical technique; validation method; visual comparison; Artificial neural networks; Demand forecasting; Energy management; Feedforward neural networks; Feedforward systems; Load forecasting; Load modeling; Neural networks; Predictive models; Zinc; Levenberg-Marquardt method; forecast; hourly load demands; neural network;

fLanguage

English

Publisher

ieee

Conference_Titel

Electrical and Computer Engineering, 2006. CCECE '06. Canadian Conference on

Conference_Location

Ottawa, Ont.

Print_ISBN

1-4244-0038-4

Electronic_ISBN

1-4244-0038-4

Type

conf

DOI

10.1109/CCECE.2006.277728

Filename

4054693